The analysis is among the first of its kind to definitively link hospital's improvement in use of guideline-recommended treatments with concomitant reductions in hospital death rates. These findings should provide compelling scientific evidence that quality improvement initiatives are worth it, and translate into significant savings in patients' lives, the researchers said.
"These findings should be a strong motivation to people, who until now found it difficult to commit to quality improvement initiatives without evidence that they work," said cardiologist Eric Peterson, M.D., who presented the results of the Duke analysis Nov. 10, 2004, at the annual scientific sessions of American Heart Association (AHA) in New Orleans. "This study shows what a profound influence quality improvement can have on saving patients lives."
The study involved analyzing reports of hospitals' adherence to treatment guidelines and mortality rates over a two-year period, from 2002 to 2003.
"When we looked at the hospitals as a group at the beginning, they were almost indistinguishable from each other in their capabilities and services offered," said Peterson "The only difference was that over time some changed their practices according to the guidelines and others did not.
"However, when we then looked at how mortality rates changed from baseline to the latest quarter, what we found was remarkable," he continued. "Those hospitals that were the worst at following the guidelines saw their mortality rates increase, while those hospitals that had the largest improvement in adherence had the greatest decrease in mortality rates. We believe this is the best argument for hospitals to devote the necessary time and effort into improving their systems for taking care of these patients."
Specifically, overtime, the mortality risks rose by 3.1 percent at hospitals whose care had worsened. In contrast, mortality risks declined by 37 percent over the same time period among hospitals whose care patterns were most improved.
Both the American College of Cardiology and the AHA have issued guidelines for optimal care of patients who arrive at hospital with symptoms of a possible heart attack, such as chest pain (unstable angina), irregular readings on an electrocardiograph or elevated chemical markers of cell death.
The guidelines, adopted after large-scale clinical trials demonstrated the effectiveness of specific therapies in saving lives, focus on giving suspected heart attack patients anti-platelet medications, heparin, glycoprotein IIb/IIIa inhibitors (clot inhibitors) or beta-blockers within the first 24 hours of admission. The guidelines also call for prescribing such drugs as aspirin, beta-blockers, ACE inhibitors or statins after discharge.
For the analysis, Peterson drew on the database of a national quality improvement initiative known as CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC and AHA Guidelines). CRUSADE maintains a national registry of data collected from more than 400 hospitals nationwide and then reports back to each hospital every three months on their adherence to the guidelines.
The current analysis focused on the quarterly changes in guideline adherence and mortality rate changes at 315 hospitals from all of 2002 through September of 2003. During that period, a total of 21,588 patients had been treated.
The hospitals were then divided into quartiles based on the degree to which they adhered to the guidelines. Each hospital was characterized as either: process worsening, no improvement, modest improvement or large improvement.
Altogether, adherence to the guidelines by the hospitals in the study improved from 67.9 percent to 77.3 percent during the study period. Over the course of analysis, the hospitals termed process-worsening had a negative 4.6 percent adherence to the guidelines and an average 3.1 percent increase in mortality. On the other end of the spectrum, the hospitals with the largest improvement had a 15.6 percent increase in adherence and a 37 percent decline in death rates.
The next step, according to Peterson, is to better understand the particular reasons behind each hospital's inability to implement quality improvement initiatives.
"Just because a hospital is participating in a program like CRUSADE is insufficient alone in making a change for the better," Peterson said. "We will be conducting in-depth interviews at different hospitals to characterize why some are refractory to change, with a goal of helping them get better."
CRUSADE is coordinated by the DCRI. It is funded by Millennium Pharmaceuticals, Cambridge, Mass., and Schering Corp, Kenilworth, N.J. Bristol-Meyers Squibb/Sanofi Pharmaceuticals Partnership, NY, provided an unrestricted grant in support of CRUSADE.
Other members of the team were Anita Chen and Matthew Roe, Duke; Michael Kontos, Medical College of Virginia, Richmond; Sidney Smith and Magnus Ohman, University of North Carolina-Chapel Hill; Chadwick Miller, Wake Forest University, Winston-Salem, N.C.; and Charles Pollack, University of Pennsylvania School of Medicine, Philadelphia.
Note to editors: The researchers involved in CRUSADE can only discuss data in the aggregate, and not about specific hospitals.